Passive Immunization against Cytomegalovirus during Pregnancy
http://www.100md.com
《新英格兰医药杂志》
To the Editor: Nigro et al. (Sept. 29 issue)1 report on the administration of cytomegalovirus (CMV) hyperimmune globulin to 68 pregnant women with primary CMV infection and conclude that this treatment "may be effective in the treatment and prevention of congenital CMV infection." However, 39 of 45 newborns with amniocentesis-proven congenital infection still shed virus at birth, and in utero infection was never documented in the remaining 84 newborns in the "prevention" group (in which women who had a recent primary infection before 21 weeks' gestation or who declined amniocentesiswere offered monthly hyperimmune globulin), only 37 of whose mothers received CMV hyperimmune globulin prenatally. Although amelioration of the severity of newborn disease was achieved, a claim of prevention of disease is not fully supported by the data.
We recently reported a case of successful clearance of in utero CMV infection in an immunosuppressed renal-transplant recipient whose CMV infection was reactivated during pregnancy.2 Serial amniocenteses confirmed initial fetal infection with subsequent clearance in conjunction with therapeutic levels of ganciclovir in the amniotic fluid. We propose that, just as antiviral agents specific to the human immunodeficiency virus (HIV) prevent vertical HIV transmission,3 CMV-specific antiviral therapies (including the preliminary findings of Nigro et al. regarding CMV immunoglobulin) be incorporated into future research protocols for preventing newborn CMV infection and illness.
Neil S. Silverman, M.D.
Center for Fetal Medicine and Women's Ultrasound
Los Angeles, CA 90048
nsilvermanmd@adelphia.net
Dechu Puliyanda, M.D.
Debbie Lehman, M.D.
Cedars-Sinai Medical Center
Los Angeles, CA 90048
References
Nigro G, Adler SP, La Torre R, Best AM. Passive immunization during pregnancy for congenital cytomegalovirus infection. N Engl J Med 2005;353:1350-1362.
Puliyanda DP, Silverman NS, Lehman D, et al. Successful use of oral ganciclovir for the treatment of intrauterine cytomegalovirus infection in a renal allograft recipient. Transpl Infect Dis 2005;7:71-74.
Watts DH. Management of human immunodeficiency virus infection in pregnancy. N Engl J Med 2002;346:1879-1891.
To the Editor: Nigro et al. recently showed that CMV-specific hyperimmune globulin may be effective in the treatment of primary CMV infection in pregnant women and the prevention of CMV in their infants. The efficacy of the treatment in the therapy group is particularly amazing in the 15 women who were carrying a fetus with evidence of fetal disease (the majority of which had intrauterine growth restriction and five of which had fetal cerebral ultrasonographic signs of disease), but the assessment would have required additional details (e.g., percentile of the birth weight).
Furthermore, the efficiency of the placental transfer of natural anti-CMV IgG antibodies is enhanced substantially as gestation progresses.1 Consequently, an accurate appraisal of the effects of various doses in the prevention of congenital CMV disease should have included quantification of viral load.2 The absence of a reduction in the rate of viremia in the group of women receiving the "prevention" treatment, in whom the time between maternal infection and hyperimmune globulin administration was shorter than in the "therapy group" (5 to 9 vs. 10 to 15 weeks), would not support the proposed "prevention" treatment.
Lionel Carbillon, M.D., Ph.D.
Assistance Publique–H?pitaux de Paris
93143 Bondy, France
lionel.carbillon@jvr.aphp.fr
References
Mussi-Pinhata MM, Pinto PC, Yamamoto AY, et al. Placental transfer of naturally acquired, maternal cytomegalovirus antibodies in term and preterm neonates. J Med Virol 2003;69:232-239.
Snydman DR. Cytomegalovirus immunoglobulins in the prevention and treatment of cytomegalovirus disease. Rev Infect Dis 1990;12:Suppl 7:S839-S848.
To the Editor: The possibility of significantly reducing the rate of intrauterine transmission of CMV and virtually abolishing the risk of congenital disease by the administration of CMV hyperimmune globulin to CMV-infected pregnant women, as reported by Nigro et al., sounds extraordinarily important. Unfortunately, the article does not provide evidence that this might be the case because it reports an uncontrolled study. Currently, women of childbearing age are neither informed about CMV nor tested for CMV antibodies. In addition, in the absence of screening during pregnancy, primary CMV infections in pregnant women are mostly undiagnosed, and no treatment is available for those with proven primary infection. In the context of such global indifference to the problem, the publication of an uncontrolled study claiming such impressive results is unhelpful. Pregnant women are a vulnerable group and deserve scientifically valid results. Investigators including Nigro and La Torre have already published similar data,1 and they should have confirmed their original results in a properly controlled study.
Maria Grazia Revello, M.D.
Istituto di Ricovero e Cura a Carattere Scientifico Policlinico
San Matteo
27100 Pavia, Italy
for the European Congenital CMV Initiative
References
Cosmi E, Mazzocco M, La Torre R, Ligi P, Sali E, Nigro G. Therapy or prevention of fetal infection by cytomegalovirus with immunoglobulin infusion in pregnant women with primary infection. Acta Biomed Ateneo Parmense 2000;71:Suppl 1:547-551.
The authors reply: Silverman et al. confuse infection and disease, and they state that CMV hyperimmune globulin was expected to terminate viral excretion. That infected fetuses would remained infected after the administration of CMV hyperimmune globulin was anticipated, because viral excretion in infected newborns persists for years regardless of neutralizing antibodies. In our prevention group, the end point was not prevention of disease but prevention of infection, which occurred. In this group, a reasonable assumption was that the uninfected infants were never infected in utero. We did not determine whether the reduced infection rate was associated with lowered maternal or placental viral loads. Ganciclovir reduces viral loads, but CMV excretion resumes after cessation of the drug, so its efficacy may be similar to that of CMV hyperimmune globulin.
To Dr. Carbillon, the effect of the administration of CMV hyperimmune globulin was "amazing," but it should not have been if hyperimmune globulin resolved fetal disease by neutralizing the virus and thus reducing placental inflammation and insufficiency, which would lead to improved fetal nutrition and oxygenation. This is plausible because most manifestations of congenital CMV infection resolve over the first weeks of life with improved nutrition and oxygenation. Our unpublished data suggest that placental size doubled throughout the second half of pregnancy among mothers with symptomatic fetuses, and the enlarged placentas decreased in size after the administration of CMV hyperimmune globulin to mothers with a primary CMV infection. Thus, one site of action of CMV hyperimmune globulin is probably the placenta, and the manifestations of congenital CMV at birth are probably caused in part by placental insufficiency.
All the newborns who had been treated with CMV hyperimmune globulin had birth weights appropriate for their gestational ages. Reduced maternal viral loads may be associated with the reduced rate of fetal infection in the prevention group. However, even if CMV hyperimmune globulin does not reduce maternal viral load, it may still prevent transmission of CMV to the fetus by reducing either placental viral load or inflammation.
In our prospective study, for ethical reasons, we were not permitted by the ethics review boards to assign controls randomly. Our multivariate analyses, however, identified no selection bias, and they also indicated that treatment with CMV hyperimmune globulin was an independent predictor of fetal outcome. Furthermore, all relevant biologic data are consistent with our observations, which included the following findings: the CMV hyperimmune globulin contained high levels of neutralizing antibody; the rates of CMV transmission and disease that we observed were in line with those previously reported by others; preexisting maternal immunity is protective against CMV infection and disease; and in experiments in animals going back 25 years, passive immunization is effective.1 We encourage Dr. Revello and her colleagues to conduct confirmatory trials.
Giovanni Nigro, M.D.
University La Sapienza
00161 Rome, Italy
nigrogio@libero.it
Stuart P. Adler, M.D.
Virginia Commonwealth University
Richmond, VA 23298
References
Bia FJ, Griffith BP, Tarsio M, Hsiung GD. Vaccination for the prevention of maternal and fetal infection with guinea pig cytomegalovirus. J Infect Dis 1980;142:732-738.
We recently reported a case of successful clearance of in utero CMV infection in an immunosuppressed renal-transplant recipient whose CMV infection was reactivated during pregnancy.2 Serial amniocenteses confirmed initial fetal infection with subsequent clearance in conjunction with therapeutic levels of ganciclovir in the amniotic fluid. We propose that, just as antiviral agents specific to the human immunodeficiency virus (HIV) prevent vertical HIV transmission,3 CMV-specific antiviral therapies (including the preliminary findings of Nigro et al. regarding CMV immunoglobulin) be incorporated into future research protocols for preventing newborn CMV infection and illness.
Neil S. Silverman, M.D.
Center for Fetal Medicine and Women's Ultrasound
Los Angeles, CA 90048
nsilvermanmd@adelphia.net
Dechu Puliyanda, M.D.
Debbie Lehman, M.D.
Cedars-Sinai Medical Center
Los Angeles, CA 90048
References
Nigro G, Adler SP, La Torre R, Best AM. Passive immunization during pregnancy for congenital cytomegalovirus infection. N Engl J Med 2005;353:1350-1362.
Puliyanda DP, Silverman NS, Lehman D, et al. Successful use of oral ganciclovir for the treatment of intrauterine cytomegalovirus infection in a renal allograft recipient. Transpl Infect Dis 2005;7:71-74.
Watts DH. Management of human immunodeficiency virus infection in pregnancy. N Engl J Med 2002;346:1879-1891.
To the Editor: Nigro et al. recently showed that CMV-specific hyperimmune globulin may be effective in the treatment of primary CMV infection in pregnant women and the prevention of CMV in their infants. The efficacy of the treatment in the therapy group is particularly amazing in the 15 women who were carrying a fetus with evidence of fetal disease (the majority of which had intrauterine growth restriction and five of which had fetal cerebral ultrasonographic signs of disease), but the assessment would have required additional details (e.g., percentile of the birth weight).
Furthermore, the efficiency of the placental transfer of natural anti-CMV IgG antibodies is enhanced substantially as gestation progresses.1 Consequently, an accurate appraisal of the effects of various doses in the prevention of congenital CMV disease should have included quantification of viral load.2 The absence of a reduction in the rate of viremia in the group of women receiving the "prevention" treatment, in whom the time between maternal infection and hyperimmune globulin administration was shorter than in the "therapy group" (5 to 9 vs. 10 to 15 weeks), would not support the proposed "prevention" treatment.
Lionel Carbillon, M.D., Ph.D.
Assistance Publique–H?pitaux de Paris
93143 Bondy, France
lionel.carbillon@jvr.aphp.fr
References
Mussi-Pinhata MM, Pinto PC, Yamamoto AY, et al. Placental transfer of naturally acquired, maternal cytomegalovirus antibodies in term and preterm neonates. J Med Virol 2003;69:232-239.
Snydman DR. Cytomegalovirus immunoglobulins in the prevention and treatment of cytomegalovirus disease. Rev Infect Dis 1990;12:Suppl 7:S839-S848.
To the Editor: The possibility of significantly reducing the rate of intrauterine transmission of CMV and virtually abolishing the risk of congenital disease by the administration of CMV hyperimmune globulin to CMV-infected pregnant women, as reported by Nigro et al., sounds extraordinarily important. Unfortunately, the article does not provide evidence that this might be the case because it reports an uncontrolled study. Currently, women of childbearing age are neither informed about CMV nor tested for CMV antibodies. In addition, in the absence of screening during pregnancy, primary CMV infections in pregnant women are mostly undiagnosed, and no treatment is available for those with proven primary infection. In the context of such global indifference to the problem, the publication of an uncontrolled study claiming such impressive results is unhelpful. Pregnant women are a vulnerable group and deserve scientifically valid results. Investigators including Nigro and La Torre have already published similar data,1 and they should have confirmed their original results in a properly controlled study.
Maria Grazia Revello, M.D.
Istituto di Ricovero e Cura a Carattere Scientifico Policlinico
San Matteo
27100 Pavia, Italy
for the European Congenital CMV Initiative
References
Cosmi E, Mazzocco M, La Torre R, Ligi P, Sali E, Nigro G. Therapy or prevention of fetal infection by cytomegalovirus with immunoglobulin infusion in pregnant women with primary infection. Acta Biomed Ateneo Parmense 2000;71:Suppl 1:547-551.
The authors reply: Silverman et al. confuse infection and disease, and they state that CMV hyperimmune globulin was expected to terminate viral excretion. That infected fetuses would remained infected after the administration of CMV hyperimmune globulin was anticipated, because viral excretion in infected newborns persists for years regardless of neutralizing antibodies. In our prevention group, the end point was not prevention of disease but prevention of infection, which occurred. In this group, a reasonable assumption was that the uninfected infants were never infected in utero. We did not determine whether the reduced infection rate was associated with lowered maternal or placental viral loads. Ganciclovir reduces viral loads, but CMV excretion resumes after cessation of the drug, so its efficacy may be similar to that of CMV hyperimmune globulin.
To Dr. Carbillon, the effect of the administration of CMV hyperimmune globulin was "amazing," but it should not have been if hyperimmune globulin resolved fetal disease by neutralizing the virus and thus reducing placental inflammation and insufficiency, which would lead to improved fetal nutrition and oxygenation. This is plausible because most manifestations of congenital CMV infection resolve over the first weeks of life with improved nutrition and oxygenation. Our unpublished data suggest that placental size doubled throughout the second half of pregnancy among mothers with symptomatic fetuses, and the enlarged placentas decreased in size after the administration of CMV hyperimmune globulin to mothers with a primary CMV infection. Thus, one site of action of CMV hyperimmune globulin is probably the placenta, and the manifestations of congenital CMV at birth are probably caused in part by placental insufficiency.
All the newborns who had been treated with CMV hyperimmune globulin had birth weights appropriate for their gestational ages. Reduced maternal viral loads may be associated with the reduced rate of fetal infection in the prevention group. However, even if CMV hyperimmune globulin does not reduce maternal viral load, it may still prevent transmission of CMV to the fetus by reducing either placental viral load or inflammation.
In our prospective study, for ethical reasons, we were not permitted by the ethics review boards to assign controls randomly. Our multivariate analyses, however, identified no selection bias, and they also indicated that treatment with CMV hyperimmune globulin was an independent predictor of fetal outcome. Furthermore, all relevant biologic data are consistent with our observations, which included the following findings: the CMV hyperimmune globulin contained high levels of neutralizing antibody; the rates of CMV transmission and disease that we observed were in line with those previously reported by others; preexisting maternal immunity is protective against CMV infection and disease; and in experiments in animals going back 25 years, passive immunization is effective.1 We encourage Dr. Revello and her colleagues to conduct confirmatory trials.
Giovanni Nigro, M.D.
University La Sapienza
00161 Rome, Italy
nigrogio@libero.it
Stuart P. Adler, M.D.
Virginia Commonwealth University
Richmond, VA 23298
References
Bia FJ, Griffith BP, Tarsio M, Hsiung GD. Vaccination for the prevention of maternal and fetal infection with guinea pig cytomegalovirus. J Infect Dis 1980;142:732-738.